Differentially Expressed Heterogeneous Overdispersion Genes Testing for Count Data.

Area of Science:

• Bioinformatics
• Genomics
• Statistical analysis

Background:

• RNA-sequencing (RNA-seq) is crucial for understanding biological systems by analyzing gene expression.
• Existing methods for detecting differentially expressed genes (DEGs) can lack power due to overdispersion and small sample sizes.
• Accurate DEG identification is vital for biological discovery and understanding cellular responses.

Approach:

• Propose a novel differential expression analysis procedure: heterogeneous overdispersion genes testing (DEHOGT).
• DEHOGT utilizes heterogeneous overdispersion modeling and a post-hoc inference procedure for robust RNA-seq read count analysis.
• Integrates sample information across conditions with a gene-wise estimation scheme for enhanced detection power.

Key Points:

• DEHOGT offers more flexible and adaptive overdispersion modeling compared to existing methods.
• Demonstrated superior performance over DESeq and EdgeR on synthetic RNA-seq data.
• Successfully applied to microglial cell RNA-seq data, identifying more DEGs under stress hormone treatments.

Conclusions:

• DEHOGT provides a powerful and adaptive approach for differential gene expression analysis in RNA-seq data.
• The method enhances the identification of biologically relevant DEGs, particularly in complex datasets.
• DEHOGT represents a significant advancement for RNA-seq data analysis in various biological research areas.